GazoPa is a new image search engine. Upload an image and the service will find similar looking items based on shape and colour. I tried uploading the bulletproof SUV image and the system came back with a credible bunch of results featuring grey SUVs, so from that point of view the thing seems to work quite well.
Interestingly they all claim to be able to deliver results from drawings and sketches, which seems to give more patchy results as you might expect. I tried out one or two of the demo examples on the home page and you get a pretty good match shape wise, but not so great in terms of actual item. The first results for the tulip sketch were golf club heads. But hey, what do you expect?
The big question I have with this type of visual search engine is, what is it good for? Riya and subsequently its Like.com offshoot were pioneers in this area, providing image search for shopping, but I’m wondering just how much of a market there is for this kind of search functionality in total? Hmm…
GazoPa is a similar image search service on the web in private beta by Hitachi. Users can search images from the web based on user’s own photo, drawings, images found on the web and keywords. GazoPa enables users to search for a similar image from characteristics such as a color or a shape extracted from an image itself. There are abundant quantities of images on the web, however many of these simply cannot be described by keywords. Since GazoPa uses image features to search other similar images, a vast range of images can be retrieved from the web.
While Gazoopa searches 50M images, Tineye.com searches over 1 billion images, and with spooky-good results. Check their demos at http://tineye.com/cool_searches
Close but no cigar – it did a great job finding a couple of similar cars from a stock photo, but when it came to identifying an unknown plant from the garden it was utterly hopeless. Maybe one day we'll get there.